The drug sensitivity and clinical outcome of human leukemias is highly dependent on their cell lineage of origin. This proposal will aim to define biochemical phenotypes with respect to drug metabolism that account for the lineage-specific response to antifolates exhibited by human leukemias. Polyglutamylation of classical and novel antifolates by the enzyme Folylpolyglutamate synthetase (FPGS) is essential to their pharmacological activity, resulting in prolonged intracellular retention and increased cytotoxicity. The proposed studies will test the hypothesis that the response of human leukemias to antifolates depend upon the expression of FPGS. We will investigate the biochemical and molecular basis for the reported clinical observation that a lineage- specific increase in FPGS activity occurs after in vivo leukemic blasts' exposure to these drugs. These studies will define the role of substrate affinity for FPGS and inhibition of key folate-metabolizing enzymes, and the effects of non-polyglutamylatable antifolates and natural folates. Similar studies with normal hematopoietic progenitors after exposure to antifolates will define the potential role of FPGS in drug selectivity. Further, changes in DNA methylation and FPGS mRNA expression in normal bone marrow cells and leukemic blasts will be investigated. To evaluate the clinical significance of these results, FPGS and polyglutamylation related parameters will be determined in clinical samples from antifolate sensitive and resistant leukemias. The clinical relevance of FPGS in antifolate response will also be tested by growth factor-induced upregulation of FPGS in resistant myeloid leukemic blasts' exposed to these agents, and after transfection of an inducible expression system encoding hFPGS to an "enzyme deficient" resistant leukemic phenotype. Overall, this proposal should provide a definitive answer to: 1. How important is FPGS in the clinical response to antifolates? 2. What are the effects of novel antifolates on leukemic blasts' FPGS expression, and do these differ in sensitive vs resistant phenotypes vs normal bone marrow progenitors? 3. Is the lineage-specific expression of FPGS an important clinical determinant of a biochemical phenotype predictor of antifolate tumor response?